
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
from datetime import datetime
from collections import Counter
import json
from sklearn.preprocessing import MinMaxScaler


file_path = 'com.samsung.shealth.activity.day_summary.20240115183382.csv'

dates = []
step_counts = []

# read the file line by line, starting from the third
# line because first two lines have headings

with open(file_path, 'r') as file:
    file.readline() 
    header = file.readline() 
    
    for line in file:
        fields = line.strip().split(',')

        step_count = fields[4]  
        date_str = fields[10]   
        
        date = datetime.strptime(date_str, '%Y-%m-%d %H:%M:%S.%f')
        
        dates.append(date)
        step_counts.append(step_count)

df = pd.DataFrame({'date': dates, 'step_count': step_counts})

df['step_count'] = pd.to_numeric(df['step_count'], errors='coerce')

df['update_date'] = df['date'].dt.strftime('%Y-%m-%d')  


# new DataFrame with unique dates and the maximum step count (there were some records earlier in
# the same day so I get max of that day to get the step count of full day)

df_max_step = df.groupby('update_date')['step_count'].max().reset_index()

date_range = pd.date_range(start='2022-05-19', end='2024-01-15', freq='D')
df_all_dates = pd.DataFrame({'date': date_range})

df_max_step['update_date'] = pd.to_datetime(df_max_step['update_date'])

# merge the two DataFrames on date to fill missing dates
df_filled = pd.merge(df_all_dates, df_max_step, left_on='date', right_on='update_date', how='left')

# fill missing step counts with zero
df_filled['step_count'] = df_filled['step_count'].fillna(0)

print("\nFilled DataFrame:")
print(df_filled[['date', 'step_count']])

Filled DataFrame:
          date  step_count
0   2022-05-19         0.0
1   2022-05-20      8543.0
2   2022-05-21      5274.0
3   2022-05-22     10995.0
4   2022-05-23      5573.0
..         ...         ...
602 2024-01-11      5648.0
603 2024-01-12      5475.0
604 2024-01-13      9965.0
605 2024-01-14      9661.0
606 2024-01-15      4631.0

[607 rows x 2 columns]


schedule_file_path = 'schedule.xlsx'  

df_schedule = pd.read_excel(schedule_file_path)

print(df_schedule)

          date  Lecture Hour
0   2022-08-15             0
1   2022-08-16             0
2   2022-08-17             0
3   2022-08-18             0
4   2022-08-19             0
..         ...           ...
518 2024-01-15             0
519 2024-01-16             0
520 2024-01-17             0
521 2024-01-18             0
522 2024-01-19             0

[523 rows x 2 columns]


with open('liked_posts.json', 'r') as file:
    json_data = json.load(file)

dates = [datetime.utcfromtimestamp(entry['timestamp']).strftime('%Y-%m-%d') for like in json_data['likes_media_likes'] for entry in like['string_list_data']]

#there are dates of every like so I count the occurences of dates to see how many posts I liked in a date
date_occurrences = dict(Counter(dates))

df_likes = pd.DataFrame(list(date_occurrences.items()), columns=['date', 'likes'])

print(df_likes)

           date  likes
0    2024-01-15     13
1    2024-01-14     27
2    2024-01-13     45
3    2024-01-12      2
4    2024-01-11     12
..          ...    ...
879  2021-07-04     13
880  2021-07-03      6
881  2021-07-02     17
882  2021-07-01      2
883  2021-06-29      5

[884 rows x 2 columns]


df_filled['date'] = pd.to_datetime(df_filled['date'])
df_likes['date'] = pd.to_datetime(df_likes['date'])
df_schedule['date'] = pd.to_datetime(df_schedule['date'])

df_merged_filled_likes = pd.merge(df_filled, df_likes, on='date', how='left')

df_merged_filled_likes_schedule = pd.merge(df_merged_filled_likes, df_schedule, on='date', how='outer')

df_merged_filled_likes_schedule = df_merged_filled_likes_schedule.drop(columns=['update_date'])

print(df_merged_filled_likes_schedule)

df_merged_filled_likes_schedule['date'] = pd.to_datetime(df_merged_filled_likes_schedule['date'])

# filter rows between '2022-08-15' and '2024-01-15' to avoid NaN values
date_range_condition = (df_merged_filled_likes_schedule['date'] >= '2022-08-15') & (df_merged_filled_likes_schedule['date'] <= '2024-01-15')
df_filtered = df_merged_filled_likes_schedule[date_range_condition]

# reset the index to start from 0
df_filtered = df_filtered.reset_index(drop=True)
print("The merged version of all available dates with information")

print(df_filtered)

          date  step_count  likes  Lecture Hour
0   2022-05-19         0.0   33.0           NaN
1   2022-05-20      8543.0   34.0           NaN
2   2022-05-21      5274.0   16.0           NaN
3   2022-05-22     10995.0    7.0           NaN
4   2022-05-23      5573.0   20.0           NaN
..         ...         ...    ...           ...
606 2024-01-15      4631.0   13.0           0.0
607 2024-01-16         NaN    NaN           0.0
608 2024-01-17         NaN    NaN           0.0
609 2024-01-18         NaN    NaN           0.0
610 2024-01-19         NaN    NaN           0.0

[611 rows x 4 columns]
The merged version of all available dates with information
          date  step_count  likes  Lecture Hour
0   2022-08-15      9180.0    7.0           0.0
1   2022-08-16       102.0   17.0           0.0
2   2022-08-17       727.0   49.0           0.0
3   2022-08-18       168.0   54.0           0.0
4   2022-08-19      2000.0   21.0           0.0
..         ...         ...    ...           ...
514 2024-01-11      5648.0   12.0           0.0
515 2024-01-12      5475.0    2.0           0.0
516 2024-01-13      9965.0   45.0           0.0
517 2024-01-14      9661.0   27.0           0.0
518 2024-01-15      4631.0   13.0           0.0

[519 rows x 4 columns]


sns.set(style="whitegrid")

fig, ax1 = plt.subplots(figsize=(12, 6))

sns.lineplot(x='date', y='step_count', data=df_filtered, ax=ax1, label='Step Count', color='blue')
ax1.set_ylabel('Step Count', color='blue')
ax1.tick_params('y', colors='blue')

ax2 = ax1.twinx()
sns.lineplot(x='date', y='likes', data=df_filtered, ax=ax2, label='Likes', color='green')
ax2.set_ylabel('Likes', color='green')
ax2.tick_params('y', colors='green')

ax1.set_xlabel('Date')

plt.title('Step Count and Likes Over Time')

plt.show()


df_normalized = df_filtered.copy()

std_dev_threshold = 2  

# Filter 'step_count'
mean_step_count = df_normalized['step_count'].mean()
std_step_count = df_normalized['step_count'].std()
df_normalized = df_normalized[(df_normalized['step_count'] >= mean_step_count - std_dev_threshold * std_step_count) &
                              (df_normalized['step_count'] <= mean_step_count + std_dev_threshold * std_step_count)]

# Filter 'likes'
mean_likes = df_normalized['likes'].mean()
std_likes = df_normalized['likes'].std()
df_normalized = df_normalized[(df_normalized['likes'] >= mean_likes - std_dev_threshold * std_likes) &
                              (df_normalized['likes'] <= mean_likes + std_dev_threshold * std_likes)]


scaler = MinMaxScaler()

df_normalized[['step_count', 'likes']] = scaler.fit_transform(df_normalized[['step_count', 'likes']])

window_size = 7  
df_normalized[['step_count', 'likes']] = df_normalized[['step_count', 'likes']].rolling(window=window_size).mean()

sns.set(style="whitegrid")

fig, ax1 = plt.subplots(figsize=(12, 6))

sns.lineplot(x='date', y='step_count', data=df_normalized, ax=ax1, label='Step Count (Smoothed)', color='blue')
ax1.set_ylabel('Step Count (Smoothed)', color='blue')
ax1.tick_params('y', colors='blue')

ax2 = ax1.twinx()
sns.lineplot(x='date', y='likes', data=df_normalized, ax=ax2, label='Likes (Smoothed)', color='green')
ax2.set_ylabel('Likes (Smoothed)', color='green')
ax2.tick_params('y', colors='green')

ax1.set_xlabel('Date')


plt.title(f'Smoothed Step Count and Likes Over Time (Excluding Outliers, {window_size}-day Rolling Average)')
plt.legend(loc='upper left')
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns

sns.set(style="whitegrid")

plt.figure(figsize=(15, 5))

plt.subplot(1, 3, 1)
sns.lineplot(x='date', y='step_count', data=df_filtered, marker='o', color='skyblue')
plt.xlabel('Date')
plt.ylabel('Step Count')
plt.title('Line Plot of Step Count Over Time')
plt.xticks(rotation=45)

plt.subplot(1, 3, 2)
sns.lineplot(x='date', y='likes', data=df_filtered, marker='o', color='salmon')
plt.xlabel('Date')
plt.ylabel('Likes')
plt.title('Line Plot of Likes Over Time')
plt.xticks(rotation=45)

plt.subplot(1, 3, 3)
sns.lineplot(x='date', y='Lecture Hour', data=df_filtered, marker='o', color='green')
plt.xlabel('Date')
plt.ylabel('Lecture Hour')
plt.title('Line Plot of Lecture Hour Over Time')
plt.xticks(rotation=45)

plt.tight_layout()
plt.show()

plt.figure(figsize=(15, 5))

plt.subplot(1, 3, 1)
sns.histplot(df_filtered['step_count'], bins=20, kde=True, color='skyblue')
plt.title('Step Count Distribution')

plt.subplot(1, 3, 2)
sns.histplot(df_filtered['likes'], bins=20, kde=True, color='salmon')
plt.title('Likes Distribution')

plt.subplot(1, 3, 3)
sns.histplot(df_filtered['Lecture Hour'], bins=20, kde=True, color='green')
plt.title('Lecture Hour Distribution')

plt.tight_layout()
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns


sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
sns.scatterplot(x='likes', y='Lecture Hour', data=df_filtered, hue='date', palette='viridis',legend=False)
plt.xlabel('Likes')
plt.ylabel('Lecture Hour')
plt.title('Scatter Plot of Likes vs Lecture Hour')
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns

sns.set(style="whitegrid")

plt.figure(figsize=(12, 6))
sns.lineplot(x='date', y='likes', data=df_filtered, label='Likes', marker='o')
sns.lineplot(x='date', y='Lecture Hour', data=df_filtered, label='Lecture Hour', marker='o')
plt.xlabel('Date')
plt.ylabel('Count')
plt.title('Line Plot of Likes and Lecture Hour Over Time')
plt.legend()
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns
import statsmodels.api as sm
import matplotlib.dates as mdates

sns.set(style="whitegrid")

# Apply smoothing using lowess from statsmodels
lowess_likes = sm.nonparametric.lowess(df_filtered['likes'], df_filtered['date'], frac=0.1)
lowess_lecture_hour = sm.nonparametric.lowess(df_filtered['Lecture Hour'], df_filtered['date'], frac=0.1)

# Convert the lowess results to a DataFrame
df_smoothed_likes = pd.DataFrame(lowess_likes, columns=['date', 'smoothed_likes'])
df_smoothed_lecture_hour = pd.DataFrame(lowess_lecture_hour, columns=['date', 'smoothed_lecture_hour'])

# Convert 'date' column to datetime type
df_smoothed_likes['date'] = pd.to_datetime(df_smoothed_likes['date'])
df_smoothed_lecture_hour['date'] = pd.to_datetime(df_smoothed_lecture_hour['date'])

plt.figure(figsize=(12, 6))
sns.lineplot(x='date', y='smoothed_likes', data=df_smoothed_likes, label='Smoothed Likes', marker='o')
sns.lineplot(x='date', y='smoothed_lecture_hour', data=df_smoothed_lecture_hour, label='Smoothed Lecture Hour', marker='o')

# Format the x-axis to show dates nicely
plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m-%d'))

plt.xlabel('Date')
plt.ylabel('Count')
plt.title('Smoothed Line Plot of Likes and Lecture Hour Over Time')
plt.legend()
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['date'] = pd.to_datetime(df_filtered['date'])
df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'step_count': 'mean',
    'likes': 'mean'
}).reset_index()

sns.set(style="whitegrid")

plt.figure(figsize=(12, 6))
sns.lineplot(x='week', y='step_count', data=df_weekly_avg, label='Average Step Count', marker='o')
plt.xlabel('Week')
plt.ylabel('Average Value')
plt.title('Weekly Average Step Count')
plt.legend(title='Category')
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'step_count': 'mean',
    'likes': 'mean'
}).reset_index()

df_weekly_avg['step_count_normalized'] = (df_weekly_avg['step_count'] - df_weekly_avg['step_count'].min()) / (
        df_weekly_avg['step_count'].max() - df_weekly_avg['step_count'].min())

df_weekly_avg['likes_normalized'] = (df_weekly_avg['likes'] - df_weekly_avg['likes'].min()) / (
        df_weekly_avg['likes'].max() - df_weekly_avg['likes'].min())

sns.set(style="whitegrid")

plt.figure(figsize=(12, 6))
sns.lineplot(x='week', y='step_count_normalized', data=df_weekly_avg, label='Normalized Step Count', marker='o')
sns.lineplot(x='week', y='likes_normalized', data=df_weekly_avg, label='Normalized Likes', marker='o')
plt.xlabel('Week')
plt.ylabel('Normalized Value')
plt.title('Normalized Weekly Average Step Count and Likes')
plt.legend(title='Category')
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns


df_filtered['date'] = pd.to_datetime(df_filtered['date'])
df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'step_count': 'mean',
    'likes': 'mean',
    'date': 'first' 
}).reset_index()


sns.set(style="whitegrid")

plt.figure(figsize=(20, 10))
sns.lineplot(x='week', y='step_count', data=df_weekly_avg, hue='year', palette='viridis', marker='o')


for i, row in df_weekly_avg.iterrows():
    plt.text(row['week'], row['step_count'], f"{row['date'].strftime('%Y-%m-%d')}", ha='center', va='bottom', rotation=45)

plt.xticks(rotation=45)
plt.xlabel('Week')
plt.ylabel('Average Value')
plt.title('Weekly Average Step Count with Dates (Separated by Year)')
plt.legend(title='Year')
plt.tight_layout()
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns


df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'step_count': 'mean',
    'likes': 'mean',
    'date': 'first' 
}).reset_index()

sns.set(style="whitegrid")

plt.figure(figsize=(12, 6))
sns.lineplot(x='week', y='step_count', data=df_weekly_avg, hue='year', palette='viridis', marker='o')

plt.xticks(rotation=45)
plt.xlabel('Week')
plt.ylabel('Average Value')
plt.title('Weekly Average Step Count (Separated by Year)')
plt.legend(title='Year')
plt.tight_layout()
plt.show()


correlation_matrix = df_filtered[['step_count', 'likes', 'Lecture Hour']].corr()

print(correlation_matrix)

sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
plt.title('Correlation Matrix')
plt.show()

              step_count     likes  Lecture Hour
step_count      1.000000 -0.238293      0.168240
likes          -0.238293  1.000000     -0.249649
Lecture Hour    0.168240 -0.249649      1.000000


df_filtered.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 519 entries, 0 to 518
Data columns (total 6 columns):
 #   Column        Non-Null Count  Dtype         
---  ------        --------------  -----         
 0   date          519 non-null    datetime64[ns]
 1   step_count    519 non-null    float64       
 2   likes         483 non-null    float64       
 3   Lecture Hour  519 non-null    float64       
 4   week          519 non-null    UInt32        
 5   year          519 non-null    UInt32        
dtypes: UInt32(2), datetime64[ns](1), float64(3)
memory usage: 21.4 KB


import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['month'] = df_filtered['date'].dt.month
df_filtered['year'] = df_filtered['date'].dt.year

monthly_avg = df_filtered.groupby('month')['step_count'].mean()
yearly_avg = df_filtered.groupby('year')['step_count'].mean()

plt.figure(figsize=(12, 6))
sns.barplot(x=monthly_avg.index, y=monthly_avg.values, color='blue')
plt.xlabel('Month')
plt.ylabel('Average Step Count')
plt.title('Monthly Average Step Count')
plt.show()

plt.figure(figsize=(12, 6))
sns.barplot(x=yearly_avg.index, y=yearly_avg.values, color='green')
plt.xlabel('Year')
plt.ylabel('Average Step Count')
plt.title('Yearly Average Step Count')
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns

sns.set(style="whitegrid")
print(df_filtered['step_count'].apply(type).unique())
print(df_filtered.head())

plt.figure(figsize=(12, 6))
sns.lineplot(x='month', y='step_count', data=df_filtered[2:], label='Step Count', marker='o')
plt.xlabel('Month')
plt.ylabel('Step Count')
plt.title('Step Count Monthly Seasonality')
plt.xticks(rotation=45)
plt.legend()
plt.show()

[<class 'float'>]
        date  step_count  likes  Lecture Hour  week  year  month
0 2022-08-15      9180.0    7.0           0.0    33  2022      8
1 2022-08-16       102.0   17.0           0.0    33  2022      8
2 2022-08-17       727.0   49.0           0.0    33  2022      8
3 2022-08-18       168.0   54.0           0.0    33  2022      8
4 2022-08-19      2000.0   21.0           0.0    33  2022      8


import seaborn as sns
import matplotlib.pyplot as plt

step_count_values = df_filtered['step_count']

sns.histplot(step_count_values, bins=20, kde=True, color='blue', edgecolor='black')

plt.xlabel('Step Count')
plt.ylabel('Frequency')
plt.title('Histogram of Step Counts')
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['month'] = df_filtered['date'].dt.month
df_filtered['year'] = df_filtered['date'].dt.year

monthly_avg_likes = df_filtered.groupby('month')['likes'].mean()
yearly_avg_likes = df_filtered.groupby('year')['likes'].mean()

plt.figure(figsize=(12, 6))
sns.barplot(x=monthly_avg_likes.index, y=monthly_avg_likes.values, color='purple')
plt.xlabel('Month')
plt.ylabel('Average Likes')
plt.title('Monthly Average Likes')
plt.show()

plt.figure(figsize=(12, 6))
sns.barplot(x=yearly_avg_likes.index, y=yearly_avg_likes.values, color='orange')
plt.xlabel('Year')
plt.ylabel('Average Likes')
plt.title('Yearly Average Likes')
plt.show()


import seaborn as sns
import matplotlib.pyplot as plt

likes_values = df_filtered['likes']

sns.histplot(likes_values, bins=20, kde=True, color='skyblue', edgecolor='black')

plt.xlabel('Likes')
plt.ylabel('Frequency')
plt.title('Histogram of Likes')
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['month'] = df_filtered['date'].dt.month

monthly_avg_lecture_hour = df_filtered.groupby('month')['Lecture Hour'].mean()

plt.figure(figsize=(12, 6))
sns.barplot(x=monthly_avg_lecture_hour.index, y=monthly_avg_lecture_hour.values, color='green')
plt.xlabel('Month')
plt.ylabel('Average Lecture Hour')
plt.title('Monthly Average Lecture Hour')
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['day_of_week'] = df_filtered['date'].dt.day_of_week

days_order = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday']
df_day_of_week_avg = df_filtered.groupby('day_of_week')['Lecture Hour'].mean().reset_index()

sns.set(style="whitegrid")
plt.figure(figsize=(10, 6))
sns.barplot(x='day_of_week', y='Lecture Hour', data=df_day_of_week_avg)
plt.xlabel('Day of the Week')
plt.ylabel('Average Lecture Hour')
plt.title('Average Lecture Hour by Day of the Week')
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['month'] = df_filtered['date'].dt.to_period('M')
df_monthly_avg = df_filtered.groupby('month').agg({
    'step_count': 'mean',
    'likes': 'mean',
    'Lecture Hour': 'mean'
}).reset_index()

def zscore_normalize(series):
    return (series - series.mean()) / series.std()

df_monthly_avg_normalized = df_monthly_avg.copy()
df_monthly_avg_normalized[['step_count', 'likes', 'Lecture Hour']] = df_monthly_avg_normalized[['step_count', 'likes', 'Lecture Hour']].apply(zscore_normalize)

df_monthly_avg_normalized_melted = pd.melt(df_monthly_avg_normalized, id_vars='month', var_name='Metric', value_name='Normalized Value')

sns.set(style="whitegrid")

# Create a bar plot for the normalized monthly averages
plt.figure(figsize=(16, 12))
sns.barplot(x='month', y='Normalized Value', hue='Metric', data=df_monthly_avg_normalized_melted)
plt.xlabel('Month')
plt.ylabel('Normalized Value')
plt.title('Normalized Monthly Averages Comparison')
plt.xticks(rotation=45)
plt.legend(title='Metric')
plt.show()


df_monthly_avg


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['month'] = df_filtered['date'].dt.to_period('M')

df_monthly_avg = df_filtered.groupby('month').agg({
    'step_count': 'mean',
    'likes': 'mean',
    'Lecture Hour': 'mean'
}).reset_index()

plt.figure(figsize=(12, 8))
sns.scatterplot(x='step_count', y='likes', size='Lecture Hour', hue='month', data=df_monthly_avg, sizes=(50, 200), alpha=0.7)
plt.xlabel('Step Count (Monthly Average)')
plt.ylabel('Likes (Monthly Average)')
plt.title('Bubble Chart: Step Count vs Likes with Lecture Hour as Bubble Size')
plt.legend(title='Month', bbox_to_anchor=(1, 1))
plt.show()


import seaborn as sns

selected_columns = ['step_count', 'likes', 'Lecture Hour']

sns.pairplot(df_filtered[selected_columns])
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.preprocessing import MinMaxScaler


df_filtered['month'] = df_filtered['date'].dt.to_period('M')

df_grouped = df_filtered.groupby('month')[['likes', 'step_count', 'Lecture Hour']].sum().reset_index()

scaler = MinMaxScaler()
df_grouped_normalized = df_grouped.copy()
df_grouped_normalized[['likes', 'step_count', 'Lecture Hour']] = scaler.fit_transform(df_grouped[['likes', 'step_count', 'Lecture Hour']])

sns.set(style="whitegrid")

plt.figure(figsize=(12, 6))
sns.barplot(x='month', y='likes', data=df_grouped_normalized, color='#3498db', label='Likes', alpha=0.7)
sns.barplot(x='month', y='step_count', data=df_grouped_normalized, color='#2ecc71', label='Step Count', alpha=0.7)
sns.barplot(x='month', y='Lecture Hour', data=df_grouped_normalized, color='#e67e22', label='Lecture Hour', alpha=0.7)

plt.xlabel('Month')
plt.ylabel('Normalized Total Sum')
plt.title('Normalized Bar Chart: Total Sum of Likes, Step Count, and Lecture Hour Over Time')
plt.xticks(rotation=45)
plt.legend(title='Category', bbox_to_anchor=(1, 1))
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'step_count': 'mean'
}).reset_index()

sns.set(style="whitegrid")

plt.figure(figsize=(12, 6))
sns.histplot(data=df_weekly_avg, x='step_count', hue='year', element='step', stat='density', common_norm=False, kde=True, palette='viridis')

plt.xlabel('Average Step Count')
plt.ylabel('Density')
plt.title('Distribution of Weekly Average Step Count (Separated by Year)')
plt.tight_layout()
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_filtered_2023 = df_filtered[df_filtered['year'] == 2023]

df_weekly_avg_2023 = df_filtered_2023.groupby(['year', 'week']).agg({
    'step_count': 'mean'
}).reset_index()

sns.set(style="whitegrid")

plt.figure(figsize=(12, 6))
sns.histplot(data=df_weekly_avg_2023, x='step_count', hue='year', element='step', stat='density', common_norm=False, kde=True, palette='viridis')

plt.xlabel('Average Step Count')
plt.ylabel('Density')
plt.title('Distribution of Weekly Average Step Count in 2023')
plt.tight_layout()
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['date'] = pd.to_datetime(df_filtered['date'])
df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg_likes = df_filtered.groupby(['year', 'week']).agg({
    'likes': 'mean'
}).reset_index()

sns.set(style="whitegrid")

plt.figure(figsize=(12, 6))
sns.histplot(data=df_weekly_avg_likes, x='likes', hue='year', element='step', stat='density', common_norm=False, kde=True, palette='viridis')

plt.xlabel('Average Likes')
plt.ylabel('Density')
plt.title('Distribution of Weekly Average Likes (Separated by Year)')
plt.tight_layout()
plt.show()


import seaborn as sns
sns.pairplot(df_filtered)
plt.show()


corr_matrix = df_filtered.corr()
sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', linewidths=0.5)
plt.show()


sns.set(style="ticks")
sns.pairplot(df_filtered, hue="Lecture Hour")
plt.show()


from mpl_toolkits.mplot3d import Axes3D
fig = plt.figure(figsize=(10, 8))
ax = fig.add_subplot(111, projection='3d')
ax.scatter(df_filtered['step_count'], df_filtered['Lecture Hour'], df_filtered['likes'], c='blue', marker='o')
ax.set_xlabel('Steps')
ax.set_ylabel('Lectures')
ax.set_zlabel('Likes')
plt.show()


sns.jointplot(x='likes', y='step_count', data=df_filtered, kind='scatter')
plt.show()


sns.jointplot(x='step_count', y='Lecture Hour', data=df_filtered, kind='scatter')
plt.show()


sns.jointplot(x='Lecture Hour', y='likes', data=df_filtered, kind='scatter')
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
import seaborn as sns


df_filtered['date'] = pd.to_datetime(df_filtered['date'])
df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'likes': 'mean',
    'step_count': 'mean'
}).reset_index()

sns.set(style="whitegrid")

fig = plt.figure(figsize=(12, 8))
ax = fig.add_subplot(111, projection='3d')

sc = ax.scatter(df_weekly_avg['week'], df_weekly_avg['year'], df_weekly_avg['step_count'],
                c=df_weekly_avg['likes'], cmap='viridis', s=50, depthshade=True)

cbar = fig.colorbar(sc, ax=ax, pad=0.1)
cbar.set_label('Average Likes', rotation=270, labelpad=15)

ax.set_xlabel('Week')
ax.set_ylabel('Year')
ax.set_zlabel('Average Step Count')
ax.set_title('3D Scatter Plot: Step Count vs Week vs Year with Likes Gradient')

plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

df_filtered['date'] = pd.to_datetime(df_filtered['date'])
df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'likes': 'mean',
    'step_count': 'mean'
}).reset_index()

plt.figure(figsize=(10, 10))

theta = np.linspace(0, 2*np.pi, len(df_weekly_avg['week']), endpoint=False)
width = 2*np.pi / len(df_weekly_avg['week'])

ax = plt.subplot(111, projection='polar')
ax.bar(theta, df_weekly_avg['likes'], width=width, alpha=0.7, label='Likes')

ax.bar(theta, df_weekly_avg['step_count'], width=width, alpha=0.7, label='Step Count')

ax.set_xticks(theta)
ax.set_xticklabels(df_weekly_avg['week'])
ax.set_title('Circular Area Chart: Likes and Step Count for Each Week')
ax.legend(loc='upper right')

plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
from sklearn.preprocessing import MinMaxScaler

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'likes': 'mean',
    'step_count': 'mean'
}).reset_index()

scaler = MinMaxScaler()
df_weekly_avg[['likes', 'step_count']] = scaler.fit_transform(df_weekly_avg[['likes', 'step_count']])

plt.figure(figsize=(10, 10))

theta = np.linspace(0, 2*np.pi, len(df_weekly_avg['week']), endpoint=False)
width = 2*np.pi / len(df_weekly_avg['week'])

ax = plt.subplot(111, projection='polar')
ax.bar(theta, df_weekly_avg['likes'], width=width, alpha=0.7, label='Likes')

ax.bar(theta, df_weekly_avg['step_count'], width=width, alpha=0.7, label='Step Count')

ax.set_xticks(theta)
ax.set_xticklabels(df_weekly_avg['week'])
ax.set_title('Normalized Circular Area Chart: Likes and Step Count for Each Week')
ax.legend(loc='upper right')

plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
from sklearn.preprocessing import MinMaxScaler


df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year

df_weekly_avg = df_filtered.groupby(['year', 'week']).agg({
    'likes': 'mean',
    'step_count': 'mean',
    'Lecture Hour': 'mean'
}).reset_index()

scaler = MinMaxScaler()
df_weekly_avg[['likes', 'step_count', 'Lecture Hour']] = scaler.fit_transform(df_weekly_avg[['likes', 'step_count', 'Lecture Hour']])

plt.figure(figsize=(10, 10))

theta = np.linspace(0, 2*np.pi, len(df_weekly_avg['week']), endpoint=False)
width = 2*np.pi / len(df_weekly_avg['week'])

ax = plt.subplot(111, projection='polar')
ax.bar(theta, df_weekly_avg['likes'], width=width, alpha=0.7, label='Likes')

ax.bar(theta, df_weekly_avg['step_count'], width=width, alpha=0.7, label='Step Count')

ax.bar(theta, df_weekly_avg['Lecture Hour'], width=width, alpha=0.7, label='Lecture Hour')

ax.set_xticks(theta)
ax.set_xticklabels(df_weekly_avg['week'])
ax.set_title('Normalized Circular Area Chart: Likes, Step Count, and Lecture Hour for Each Week')
ax.legend(loc='upper right')

plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['day_of_week'] = df_filtered['date'].dt.day_name()

df_day_of_week_total_likes = df_filtered.groupby('day_of_week')['likes'].sum().reset_index()

day_order = ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday"]

sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
sns.barplot(x='day_of_week', y='likes', data=df_day_of_week_total_likes, order=day_order)
plt.xlabel('Day of the Week')
plt.ylabel('Total Likes')
plt.title('Total Likes for Each Day of the Week')
plt.show()


df_filtered


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
sns.barplot(x='day_of_week', y='likes', data=df_filtered, order=day_order)
plt.xlabel('Day of the Week')
plt.ylabel('Average Likes')
plt.title('Average Likes for Each Day of the Week')
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['day_of_week'] = df_filtered['date'].dt.dayofweek

df_filtered['day_type'] = df_filtered['day_of_week'].apply(lambda x: 'Weekend' if x >= 5 else 'Weekday')

sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
sns.barplot(x='day_type', y='likes', data=df_filtered)
plt.xlabel('Day Type')
plt.ylabel('Average Likes')
plt.title('Average Likes for Weekdays vs Weekends')
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns


df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['day_of_week'] = df_filtered['date'].dt.dayofweek

df_weekday_total_steps = df_filtered.groupby('day_of_week')['step_count'].sum().reset_index()

sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
sns.barplot(x='day_of_week', y='step_count', data=df_weekday_total_steps)
plt.xlabel('Day of the Week')
plt.ylabel('Total Step Count')
plt.title('Total Step Count for Each Weekday')
plt.xticks(ticks=range(7), labels=['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday'])
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns


df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['day_of_week'] = df_filtered['date'].dt.dayofweek

df_weekday_avg_steps = df_filtered.groupby('day_of_week')['step_count'].mean().reset_index()

sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
sns.barplot(x='day_of_week', y='step_count', data=df_weekday_avg_steps)
plt.xlabel('Day of the Week')
plt.ylabel('Average Step Count')
plt.title('Average Step Count for Each Weekday')
plt.xticks(ticks=range(7), labels=['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday'])
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['day_of_week'] = df_filtered['date'].dt.dayofweek

df_filtered['day_type'] = df_filtered['day_of_week'].apply(lambda x: 'Weekend' if x >= 5 else 'Weekday')

sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
sns.barplot(x='day_type', y='step_count', data=df_filtered)
plt.xlabel('Day Type')
plt.ylabel('Average Step Count')
plt.title('Average Step Count for Weekdays vs Weekends')
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns


df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['day_of_week'] = df_filtered['date'].dt.dayofweek

sns.set(style="whitegrid")

plt.figure(figsize=(10, 6))
sns.barplot(x='day_of_week', y='Lecture Hour', data=df_filtered, ci=None)  # ci=None removes error bars
plt.xlabel('Day of the Week')
plt.ylabel('Average Lecture Hour')
plt.title('Average Lecture Hour for Each Day of the Week')
plt.xticks(range(7), ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday'])
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns


df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['day_of_week'] = df_filtered['date'].dt.dayofweek

sns.set(style="whitegrid")

plt.figure(figsize=(8, 10))

plt.subplot(3, 1, 1)
sns.barplot(x='day_of_week', y='step_count', data=df_filtered, ci=None)
plt.xlabel('Day of the Week')
plt.ylabel('Average Step Count')
plt.title('Average Step Count for Each Day of the Week')

plt.subplot(3, 1, 2)
sns.barplot(x='day_of_week', y='likes', data=df_filtered, ci=None)
plt.xlabel('Day of the Week')
plt.ylabel('Average Likes')
plt.title('Average Likes for Each Day of the Week')

plt.subplot(3, 1, 3)
sns.barplot(x='day_of_week', y='Lecture Hour', data=df_filtered, ci=None)
plt.xlabel('Day of the Week')
plt.ylabel('Average Lecture Hour')
plt.title('Average Lecture Hour for Each Day of the Week')

plt.tight_layout()
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.preprocessing import MinMaxScaler

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['day_of_week'] = df_filtered['date'].dt.dayofweek

sns.set(style="whitegrid")

df_weekdays = df_filtered[df_filtered['day_of_week'] < 8]

scaler = MinMaxScaler()
df_normalized = df_weekdays.copy()
df_normalized[['likes', 'step_count', 'Lecture Hour']] = scaler.fit_transform(df_normalized[['likes', 'step_count', 'Lecture Hour']])

plt.figure(figsize=(12, 8))

sns.lineplot(x='day_of_week', y='likes', data=df_normalized, ci=None, color='blue', marker='o', label='Likes')
sns.lineplot(x='day_of_week', y='step_count', data=df_normalized, ci=None, color='red', marker='s', label='Step Count')
sns.lineplot(x='day_of_week', y='Lecture Hour', data=df_normalized, ci=None, color='green', marker='s', label='Lecture Hour')

plt.ylim(-0.1, 0.5)

plt.xlabel('Day of the Week')
plt.ylabel('Normalized Value')
plt.title('Normalized Likes, Step Count, and Lecture Hour for Each Weekday')

plt.legend(loc='upper right')

plt.tight_layout()
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.preprocessing import MinMaxScaler

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['month'] = df_filtered['date'].dt.month
df_filtered['year'] = df_filtered['date'].dt.year

df_2023 = df_filtered[(df_filtered['year'] == 2023) & (df_filtered['day_of_week'] < 8)]

scaler = MinMaxScaler()
df_normalized = df_2023.copy()
df_normalized[['likes', 'step_count', 'Lecture Hour']] = scaler.fit_transform(df_normalized[['likes', 'step_count', 'Lecture Hour']])

plt.figure(figsize=(12, 8))

sns.lineplot(x='month', y='likes', data=df_normalized, ci=None, color='blue', marker='o', label='Likes')
sns.lineplot(x='month', y='step_count', data=df_normalized, ci=None, color='red', marker='s', label='Step Count')
sns.lineplot(x='month', y='Lecture Hour', data=df_normalized, ci=None, color='green', marker='s', label='Lecture Hour')

plt.ylim(-0.1, 0.6)

plt.xlabel('Month')
plt.ylabel('Normalized Value')
plt.title('Normalized Likes, Step Count, and Lecture Hour for Each Month in 2023')

plt.legend(loc='upper right')

plt.tight_layout()
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.preprocessing import MinMaxScaler

df_filtered['date'] = pd.to_datetime(df_filtered['date'])

df_filtered['week'] = df_filtered['date'].dt.isocalendar().week
df_filtered['year'] = df_filtered['date'].dt.isocalendar().year
df_filtered['day_of_week'] = df_filtered['date'].dt.dayofweek

scaler = MinMaxScaler()
df_normalized = df_filtered.copy()
df_normalized[['likes', 'step_count', 'Lecture Hour']] = scaler.fit_transform(df_normalized[['likes', 'step_count', 'Lecture Hour']])

plt.figure(figsize=(16, 8))

sns.lineplot(x='week', y='likes', data=df_normalized, ci=None, color='blue', marker='o', label='Likes')
sns.lineplot(x='week', y='step_count', data=df_normalized, ci=None, color='red', marker='s', label='Step Count')
sns.lineplot(x='week', y='Lecture Hour', data=df_normalized, ci=None, color='green', marker='s', label='Lecture Hour')

plt.ylim(-0.1, 0.7)

plt.xlabel('Week')
plt.ylabel('Normalized Value')
plt.title('Normalized Likes, Step Count, and Lecture Hour for Each Week in the Year')

plt.legend(loc='upper right')

plt.tight_layout()
plt.show()


import pandas as pd
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler

features_for_clustering = df_filtered[['Lecture Hour', 'step_count']]

scaler = StandardScaler()
scaled_features = scaler.fit_transform(features_for_clustering)

k = 3
kmeans = KMeans(n_clusters=k, random_state=42)
df_filtered['cluster'] = kmeans.fit_predict(scaled_features)

plt.scatter(df_filtered['Lecture Hour'], df_filtered['step_count'], c=df_filtered['cluster'], cmap='viridis', marker='o')
plt.xlabel('Lecture Hour')
plt.ylabel('Step Count')
plt.title('Clustering of Data Points')
plt.show()


df_filtered


import pandas as pd

df_filtered['interaction_term'] = df_filtered['Lecture Hour'] * df_filtered['step_count']

df_filtered.isnull()


import statsmodels.api as sm

X = sm.add_constant(df_filtered[['Lecture Hour', 'step_count', 'interaction_term']])
y = df_filtered['likes']

model = sm.OLS(y, X).fit()

print(model.summary())

                            OLS Regression Results                            
==============================================================================
Dep. Variable:                  likes   R-squared:                         nan
Model:                            OLS   Adj. R-squared:                    nan
Method:                 Least Squares   F-statistic:                       nan
Date:                Thu, 18 Jan 2024   Prob (F-statistic):                nan
Time:                        23:11:01   Log-Likelihood:                    nan
No. Observations:                 519   AIC:                               nan
Df Residuals:                     515   BIC:                               nan
Df Model:                           3                                         
Covariance Type:            nonrobust                                         
====================================================================================
                       coef    std err          t      P>|t|      [0.025      0.975]
------------------------------------------------------------------------------------
const                   nan        nan        nan        nan         nan         nan
Lecture Hour            nan        nan        nan        nan         nan         nan
step_count              nan        nan        nan        nan         nan         nan
interaction_term        nan        nan        nan        nan         nan         nan
==============================================================================
Omnibus:                          nan   Durbin-Watson:                     nan
Prob(Omnibus):                    nan   Jarque-Bera (JB):                  nan
Skew:                             nan   Prob(JB):                          nan
Kurtosis:                         nan   Cond. No.                     4.91e+04
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The condition number is large, 4.91e+04. This might indicate that there are
strong multicollinearity or other numerical problems.

C:\ProgramData\Anaconda3\lib\site-packages\statsmodels\tsa\tsatools.py:142: FutureWarning: In a future version of pandas all arguments of concat except for the argument 'objs' will be keyword-only
  x = pd.concat(x[::order], 1)


import pandas as pd
from sklearn.model_selection import train_test_split

df_shuffled = df_filtered.sample(frac=1, random_state=42)

y_column = 'likes'
X_columns = df_shuffled.drop(columns=[y_column]).columns.tolist()

X = df_shuffled[X_columns]
y = df_shuffled[y_column]

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

print(f"X_train shape: {X_train.shape}, y_train shape: {y_train.shape}")
print(f"X_test shape: {X_test.shape}, y_test shape: {y_test.shape}")

X_train shape: (415, 10), y_train shape: (415,)
X_test shape: (104, 10), y_test shape: (104,)


import seaborn as sns
import matplotlib.pyplot as plt
df_filtered['Social_Media_Engagement_Ratio'] = df_filtered['likes'] / df_filtered['step_count']

correlations = df_filtered.corr()

plt.figure(figsize=(12, 10))
sns.heatmap(correlations, annot=True, cmap='coolwarm', fmt=".2f", linewidths=.5)
plt.title('Correlation Heatmap')
plt.show()


selected_features = correlations['likes'][(correlations['likes'] > 0.3) | (correlations['likes'] < -0.3)].index.tolist()

print("Selected Features:")
print(selected_features)

Selected Features:
['likes', 'Social_Media_Engagement_Ratio']


X_train_numeric = X_train.select_dtypes(include=['float64', 'int64'])

print("Numeric Columns:")
print(X_train_numeric.columns)

Numeric Columns:
Index(['step_count', 'Lecture Hour', 'month', 'day_of_week',
       'interaction_term'],
      dtype='object')


print("Shape of X_train_numeric:", X_train_numeric.shape)
print("Shape of y_train:", y_train.shape)

Shape of X_train_numeric: (415, 5)
Shape of y_train: (415,)


Xy_train = pd.concat([X_train_numeric, y_train], axis=1)

Xy_train.replace([np.inf, -np.inf], np.nan, inplace=True)

Xy_train_cleaned = Xy_train.dropna()

X_train_numeric_cleaned = Xy_train_cleaned.drop(columns=['likes'])
y_train_cleaned = Xy_train_cleaned['likes']

print(f"Original X_train_numeric shape: {X_train_numeric.shape}")
print(f"Cleaned X_train_numeric shape: {X_train_numeric_cleaned.shape}")
print(f"Original y_train shape: {y_train.shape}")
print(f"Cleaned y_train shape: {y_train_cleaned.shape}")

X_train, X_test, y_train, y_test = train_test_split(X_train_numeric_cleaned, y_train_cleaned, test_size=0.2, random_state=42)

Original X_train_numeric shape: (415, 5)
Cleaned X_train_numeric shape: (386, 5)
Original y_train shape: (415,)
Cleaned y_train shape: (386,)


from sklearn.tree import DecisionTreeRegressor
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error

best_dtree = DecisionTreeRegressor(max_depth=5, min_samples_split=10, random_state=42)

best_dtree.fit(X_train, y_train)

y_pred = best_dtree.predict(X_test)

mse = mean_squared_error(y_test, y_pred)
print(f"Mean Squared Error on the test set: {mse}")

Mean Squared Error on the test set: 169.1044078130208


import matplotlib.pyplot as plt
import seaborn as sns

plt.figure(figsize=(10, 6))
sns.scatterplot(x=y_test, y=y_pred)
plt.title('Actual vs. Predicted Likes')
plt.xlabel('Actual Likes')
plt.ylabel('Predicted Likes')
plt.show()


from sklearn.tree import DecisionTreeRegressor
from sklearn.metrics import mean_squared_error, r2_score

tree_model = DecisionTreeRegressor(random_state=42)

tree_model.fit(X_train, y_train)

y_pred = tree_model.predict(X_test)

mse = mean_squared_error(y_test, y_pred)
r2 = r2_score(y_test, y_pred)

print(f"Mean Squared Error: {mse:.2f}")
print(f"R-squared: {r2:.2f}")

Mean Squared Error: 426.44
R-squared: -1.24


from sklearn.ensemble import RandomForestRegressor

rf_model = RandomForestRegressor(n_estimators=100, random_state=42)

rf_model.fit(X_train, y_train)

y_pred_rf = rf_model.predict(X_test)

mse_rf = mean_squared_error(y_test, y_pred_rf)
r2_rf = r2_score(y_test, y_pred_rf)

print(f"Random Forest Mean Squared Error: {mse_rf:.2f}")
print(f"Random Forest R-squared: {r2_rf:.2f}")

Random Forest Mean Squared Error: 174.45
Random Forest R-squared: 0.08


import matplotlib.pyplot as plt
import seaborn as sns

plt.figure(figsize=(10, 6))
sns.scatterplot(x=y_test, y=y_pred)
plt.title('Actual vs. Predicted Likes')
plt.xlabel('Actual Likes')
plt.ylabel('Predicted Likes')
plt.show()


import matplotlib.pyplot as plt
import seaborn as sns

plt.figure(figsize=(10, 6))
sns.scatterplot(x=y_test, y=y_pred_rf)
plt.title('Actual vs. Predicted Likes Random Forest')
plt.xlabel('Actual Likes')
plt.ylabel('Predicted Likes')
plt.show()


import pandas as pd

# Calculate average step count based on the day of the week
day_of_week_avg_step_count = df_filtered.groupby('day_of_week')['step_count'].mean()

# Create new_week_data DataFrame
new_week_data = pd.DataFrame({'Lecture Hour': [5, 4, 4, 2, 5, 0, 0]})

# Fill in additional information
new_week_data['average_step_count'] = new_week_data['Lecture Hour'].map(day_of_week_avg_step_count)
new_week_data['month'] = df_filtered['month'].mean()
new_week_data['interaction_term'] = new_week_data['Lecture Hour'] * new_week_data['average_step_count']

# Map day_of_week based on the given order
day_of_week_mapping = {0: 0, 1: 1, 2: 2, 3: 3, 4: 4, 5: 5, 6: 6}
new_week_data['day_of_week'] = new_week_data['Lecture Hour'].map(day_of_week_mapping)

# Display the resulting DataFrame
print(new_week_data)

   Lecture Hour  average_step_count     month  interaction_term  day_of_week
0             5         5612.567568  7.350674      28062.837838            5
1             4         5433.932432  7.350674      21735.729730            4
2             4         5433.932432  7.350674      21735.729730            4
3             2         6508.621622  7.350674      13017.243243            2
4             5         5612.567568  7.350674      28062.837838            5
5             0         6588.973333  7.350674          0.000000            0
6             0         6588.973333  7.350674          0.000000            0


predicted_likes_rf = rf_model.predict(new_week_data)
predicted_likes_dt = tree_model.predict(new_week_data)


print(predicted_likes_rf)
print(predicted_likes_dt)

[12.61116667 12.61116667 12.61116667 12.61116667 12.61116667 12.9945
 12.9945    ]
[5. 5. 5. 5. 5. 5. 5.]

	month	step_count	likes	Lecture Hour
0	2022-08	3480.411765	19.058824	0.000000
1	2022-09	3677.500000	18.066667	0.000000
2	2022-10	9285.516129	18.709677	2.580645
3	2022-11	7092.700000	17.583333	2.866667
4	2022-12	7948.419355	16.032258	2.612903
5	2023-01	5102.290323	28.433333	0.612903
6	2023-02	5846.678571	21.653846	0.285714
7	2023-03	5749.064516	27.354839	3.451613
8	2023-04	4405.033333	13.263158	3.066667
9	2023-05	4825.290323	14.727273	2.967742
10	2023-06	5614.333333	19.840000	0.000000
11	2023-07	3323.129032	25.903226	0.000000
12	2023-08	2903.903226	27.258065	0.000000
13	2023-09	7164.666667	13.931034	0.000000
14	2023-10	8988.838710	12.200000	3.451613
15	2023-11	8240.733333	15.600000	3.700000
16	2023-12	3486.225806	11.193548	3.451613
17	2024-01	7875.933333	14.466667	1.666667

Step Counts¶

schedule¶

instagram likes¶

Merging the Data Frames and filtering the common dates¶

visualization¶

Time Series Analysis¶

step count¶

instagram likes¶

lecture hours¶

compare monthly¶

EDA Multivariate¶

Machine Learning¶

	date	step_count	likes	Lecture Hour	week	year	month	day_of_week
0	2022-08-15	9180.0	7.0	0.0	33	2022	2022-08	Monday
1	2022-08-16	102.0	17.0	0.0	33	2022	2022-08	Tuesday
2	2022-08-17	727.0	49.0	0.0	33	2022	2022-08	Wednesday
3	2022-08-18	168.0	54.0	0.0	33	2022	2022-08	Thursday
4	2022-08-19	2000.0	21.0	0.0	33	2022	2022-08	Friday
...	...	...	...	...	...	...	...	...
514	2024-01-11	5648.0	12.0	0.0	2	2024	2024-01	Thursday
515	2024-01-12	5475.0	2.0	0.0	2	2024	2024-01	Friday
516	2024-01-13	9965.0	45.0	0.0	2	2024	2024-01	Saturday
517	2024-01-14	9661.0	27.0	0.0	2	2024	2024-01	Sunday
518	2024-01-15	4631.0	13.0	0.0	3	2024	2024-01	Monday

	date	step_count	likes	Lecture Hour	week	year	month	day_of_week	day_type	cluster	interaction_term
0	False	False	False	False	False	False	False	False	False	False	False
1	False	False	False	False	False	False	False	False	False	False	False
2	False	False	False	False	False	False	False	False	False	False	False
3	False	False	False	False	False	False	False	False	False	False	False
4	False	False	False	False	False	False	False	False	False	False	False
...	...	...	...	...	...	...	...	...	...	...	...
514	False	False	False	False	False	False	False	False	False	False	False
515	False	False	False	False	False	False	False	False	False	False	False
516	False	False	False	False	False	False	False	False	False	False	False
517	False	False	False	False	False	False	False	False	False	False	False
518	False	False	False	False	False	False	False	False	False	False	False